Skip to content

[jaspRegression] Add a footnote to Ordinal GLM output to clarify parameterization...#481

Open
sisyphus-jasp wants to merge 1 commit intojasp-stats:masterfrom
sisyphus-jasp:fix-sisyphus-1772853796
Open

[jaspRegression] Add a footnote to Ordinal GLM output to clarify parameterization...#481
sisyphus-jasp wants to merge 1 commit intojasp-stats:masterfrom
sisyphus-jasp:fix-sisyphus-1772853796

Conversation

@sisyphus-jasp
Copy link
Copy Markdown

Summary

Fixes: jasp-stats/jasp-issues#3967

Root cause

  • Ordinal GLM coefficient footnote was incomplete: listed levels/predictors but not how to interpret coefficient sign under the cumulative-logit parameterization.
  • Same footnote block built level labels with length(dv) (always 1), so numbering was wrong.

What changed

  • Updated ordinal-logistic footnote logic in R/generalizedlinearmodel.R (.glmEstimatesTableFill).
  • Fixed level indexing to use actual number of levels via seq_along(levels(...)).
  • Expanded footnote text to clarify parameterization: model uses logit(P(Y<=k)); positive coefficients imply higher odds of lower levels, negative coefficients imply higher odds of higher levels.

Caveats / reviewer checks

  • Only footnote text path changed; no model fitting/statistics changed.
  • Verify wording is acceptable for translators and user docs.
Implementation Plan

Root cause

  • Ordinal GLM coeff footnote in R/generalizedlinearmodel.R (.glmEstimatesTableFill) lists levels + predictor labels only; parameterization meaning not explicit, so sign interpretation unclear.
  • Same block has level-index bug: length(dv) used instead of number of factor levels, so labels render 1:<level> repeatedly.

Proposed changes

  • Edit only ordinal-logistic footnote block in .glmEstimatesTableFill.
  • Build level labels with seq_along(levels(dataset[[dv]])).
  • Expand footnote to explain cumulative-logit parameterization from this model (logit(P(Y<=k))) and sign meaning (positive -> lower levels; negative -> higher levels).
  • Keep all user text in gettextf().

Expected test impact

  • Fit/statistics unchanged.
  • Footnote text changes for ordinal GLM output; snapshot/text expectations may need refresh.
  • Non-ordinal GLM paths unaffected.

Add a footnote to Ordinal GLM output to clarify parameterization interpretation

Test Results

Test Run Result
Baseline (pre-fix) [ FAIL 1 | WARN 58 | SKIP 1 | PASS 581 ]
Post-fix [ FAIL 1 | WARN 58 | SKIP 1 | PASS 581 ]
Upstream CI 83f5ccf -- CI: failing

Automated Code Review

Approved after 1 review iteration(s).

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

[Ordinal GLM]: add footnote to clarify parameterization

2 participants